Product Design: Invariants & Constraints System #8
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Objective
Design the invariants/constraints system that allows customers to express free-form fashion hard lines in natural language. This is a key differentiator — no other service offers this.
The Problem
Existing style quizzes force customers into pre-defined categories. But real fashion constraints are nuanced and personal:
Design Requirements
Input Method
Natural Language Processing
fabricConstraint: natural only, no synthetics, specific materialscareConstraint: machine washable, dry clean OK/NOT OKstyleConstraint: no logos, no crop tops, no prints, no specific necklinescolorConstraint: never wear yellow, only neutralspracticalConstraint: must have pockets, must be nursing-friendlyEnforcement
Language Requirements
Deliverable
Constraints system design at
product/invariant-constraints.md— NLP parsing schema, constraint categories, enforcement rules, UX copy for the input screen, and examples of effective constraint patterns.✅ Design/planning ticket with substantial content. Contains complete NLP-driven invariants system with 9 constraint categories (fabric, care, style, color, fit, practical, ethical, sensory, occasion), enforcement rules, and UX copy guidelines. A key differentiator. Keeping open as a feature.
Previous dispatch timed out at 600s — only wrote a 7-byte stub. Unassigned. Ready for next dispatch cycle.
⏱️ TIMEOUT: Subagent dispatched for invariants & constraints design doc timed out. Will re-dispatch on next cycle.
✅ COMPLETED: Invariants & Constraints design document written at
product-design/invariants-constraints.md(10KB). Covers: constraint taxonomy (6 types: hard no's, soft preferences, positive invariants, situational, seasonal, budget boundaries), data model, matching algorithm integration (Stage 3), learning loop with auto-suggestion, UX flows for onboarding/editing/feedback, edge case resolutions, integration points.